WITHDRAWN: Image processing system based on FPGA and convolutional neural network

Abstract

Lung cancer is one of the deadliest diseases leading to a high mortality rate worldwide. Pulmonary diseases-based lung Cancer which is an abnormal growth of cells that can be characterized by a single irregular cell and spread to the entire lungs. Therefore, it is necessary to detect affected area and following application steps are to be adopted to find and cure it in the early stages. Lung cancer is often considered as a key indicator in the diagnosis of obstructive pulmonary disease. In the previous method, SVM (Space Vector Modulation) and STFT (Short-Time Fourier Transform algorithms) were used to process the lung cancer detection based image processing system in which CT (Computerized Tomography) images exhibit less accuracy and less efficiency. The transforming method delivers significantly slower results in processing and the image cannot be verified in advanced risk architecture. This proposed FPGA (Field-Programmable Gate Array) and CNN (Conventional Neural Network) are used to develop image processing and easily interface with data without any complexity. FPGA (Field-Programmable Gate Array) is mainly realized by ASIC (Application-Specific Integrated Circuit). This system accelerates the detection of lung and pulmonary disease detection and can be used as a single-process system or as an integral part of another biomedical image detection system. The image processing system relies on bilateral filtration, edge detection, multi-threshold, image segmentation, morphological image processing, and image labeling to collect lung cancer symptoms according to the neural network and gate array.